Tweak Ploopy predator-prey models

[dantaeyoung]

Currently Ploopy (heh) (Ploopy.py in the predator_prey folder) runs away - populations run out of hand.

This could be solved by:

• creating negative feedback loops (e.g. "rabbits->rabbits" : "-0.3")
• adding some dampening function/normalizing function that, say, sums the values of all nodes within a graph to be 1. (e.g. "rabbits" + "foxes" = 1).
• testing out a min/max range for results (e.g. setting negative values to 0, or clamping all node values to between 1 and -1.
• Consider implementing weights akin to the Lotka-Volterra equations (https://en.wikipedia.org/wiki/Lotka%E2%80%93Volterra_equations)
• Modify Ploopy not to actually calculate a simulation over time, but to calculate a equation/graph of phase space that periodically oscillates between populations. Phase space could be represented as a sequence of points through which a curve is interpolated through. Visualization will simply 'scan' across the phase space/curve.

(http://www.phschool.com/atschool/phbio/active_art/predator_prey_simulation/)

[dantaeyoung]

Phase space idea:

[dantaeyoung]

Almost implemented with lvsolver.py w/ commit 0f85358

Lvsolver needs to be implemented in World.